This PSCI hook is similar to pwr_domain_on_finish but is
guaranteed to be invoked with the respective core and cluster are
participating in coherency. This will be necessary to safely invoke
the new GICv3 API which modifies shared GIC data structures concurrently.
Change-Id: I8e54f05c9d4ef5712184c9c18ba45ac97a29eb7a
Signed-off-by: Madhukar Pappireddy <madhukar.pappireddy@arm.com>
This patch provides the following features and makes modifications
listed below:
- Individual APIAKey key generation for each CPU.
- New key generation on every BL31 warm boot and TSP CPU On event.
- Per-CPU storage of APIAKey added in percpu_data[]
of cpu_data structure.
- `plat_init_apiakey()` function replaced with `plat_init_apkey()`
which returns 128-bit value and uses Generic timer physical counter
value to increase the randomness of the generated key.
The new function can be used for generation of all ARMv8.3-PAuth keys
- ARMv8.3-PAuth specific code placed in `lib\extensions\pauth`.
- New `pauth_init_enable_el1()` and `pauth_init_enable_el3()` functions
generate, program and enable APIAKey_EL1 for EL1 and EL3 respectively;
pauth_disable_el1()` and `pauth_disable_el3()` functions disable
PAuth for EL1 and EL3 respectively;
`pauth_load_bl31_apiakey()` loads saved per-CPU APIAKey_EL1 from
cpu-data structure.
- Combined `save_gp_pauth_registers()` function replaces calls to
`save_gp_registers()` and `pauth_context_save()`;
`restore_gp_pauth_registers()` replaces `pauth_context_restore()`
and `restore_gp_registers()` calls.
- `restore_gp_registers_eret()` function removed with corresponding
code placed in `el3_exit()`.
- Fixed the issue when `pauth_t pauth_ctx` structure allocated space
for 12 uint64_t PAuth registers instead of 10 by removal of macro
CTX_PACGAKEY_END from `include/lib/el3_runtime/aarch64/context.h`
and assigning its value to CTX_PAUTH_REGS_END.
- Use of MODE_SP_ELX and MODE_SP_EL0 macro definitions
in `msr spsel` instruction instead of hard-coded values.
- Changes in documentation related to ARMv8.3-PAuth and ARMv8.5-BTI.
Change-Id: Id18b81cc46f52a783a7e6a09b9f149b6ce803211
Signed-off-by: Alexei Fedorov <Alexei.Fedorov@arm.com>
cpu_idx is used as mix of `unsigned int` and `signed int` in code
with typecasting at some places. This change is to unify the
cpu_idx as `unsigned int` as underlying API;s `plat_my_core_pos`
returns `unsigned int`
It was discovered via coverity issue CID 354715
Signed-off-by: Deepika Bhavnani <deepika.bhavnani@arm.com>
Change-Id: I4f0adb0c596ff1177210c5fe803bff853f2e54ce
GCC diagnostics were added to ignore array boundaries, instead
of ignoring GCC warning current code will check for array boundaries
and perform and array update only for valid elements.
Resolves: `CID 246574` `CID 246710` `CID 246651`
Signed-off-by: Deepika Bhavnani <deepika.bhavnani@arm.com>
Change-Id: I7530ecf7a1707351c6ee87e90cc3d33574088f57
NOTE: AARCH32/AARCH64 macros are now deprecated in favor of __aarch64__.
All common C compilers pre-define the same macros to signal which
architecture the code is being compiled for: __arm__ for AArch32 (or
earlier versions) and __aarch64__ for AArch64. There's no need for TF-A
to define its own custom macros for this. In order to unify code with
the export headers (which use __aarch64__ to avoid another dependency),
let's deprecate the AARCH32 and AARCH64 macros and switch the code base
over to the pre-defined standard macro. (Since it is somewhat
unintuitive that __arm__ only means AArch32, let's standardize on only
using __aarch64__.)
Change-Id: Ic77de4b052297d77f38fc95f95f65a8ee70cf200
Signed-off-by: Julius Werner <jwerner@chromium.org>
When acquiring or releasing the power domain locks for a given CPU the
parent nodes are looked up by walking the up the PD tree list on both the
acquire and release path, only one set of lookups is needed. Fetch the
parent nodes first and pass this list into both the acquire and release
functions to avoid the double lookup.
This also allows us to not have to do this lookup after coherency has
been exited during the core power down sequence. The shared struct
psci_cpu_pd_nodes is not placed in coherent memory like is done
for psci_non_cpu_pd_nodes and doing so would negatively affect
performance. With this patch we remove the need to have it in coherent
memory by moving the access out of psci_release_pwr_domain_locks().
Signed-off-by: Andrew F. Davis <afd@ti.com>
Change-Id: I7b9cfa9d31148dea0f5e21091c8b45ef7fe4c4ab
Enforce full include path for includes. Deprecate old paths.
The following folders inside include/lib have been left unchanged:
- include/lib/cpus/${ARCH}
- include/lib/el3_runtime/${ARCH}
The reason for this change is that having a global namespace for
includes isn't a good idea. It defeats one of the advantages of having
folders and it introduces problems that are sometimes subtle (because
you may not know the header you are actually including if there are two
of them).
For example, this patch had to be created because two headers were
called the same way: e0ea0928d5 ("Fix gpio includes of mt8173 platform
to avoid collision."). More recently, this patch has had similar
problems: 46f9b2c3a2 ("drivers: add tzc380 support").
This problem was introduced in commit 4ecca33988 ("Move include and
source files to logical locations"). At that time, there weren't too
many headers so it wasn't a real issue. However, time has shown that
this creates problems.
Platforms that want to preserve the way they include headers may add the
removed paths to PLAT_INCLUDES, but this is discouraged.
Change-Id: I39dc53ed98f9e297a5966e723d1936d6ccf2fc8f
Signed-off-by: Antonio Nino Diaz <antonio.ninodiaz@arm.com>
Initial Spectre Variant 1 mitigations (CVE-2017-5753).
A potential speculative data leak was found in PSCI code, this depends
on a non-robust implementation of the `plat_get_core_pos_by_mpidr()`
function. This is considered very low-risk. This patch adds a macro to
mitigate this. Note not all code paths could be analyzed with current
tools.
Add a macro which makes a variable 'speculation safe', using the
__builtin_speculation_safe_value function of GCC and llvm. This will be
available in GCC 9, and is planned for llvm, but is not currently in
mainline GCC or llvm. In order to implement this mitigation the compiler
must support this builtin. Support is indicated by the
__HAVE_SPECULATION_SAFE_VALUE flag.
The -mtrack-speculation option maintains a 'tracker' register, which
determines if the processor is in false speculation at any point. This
adds instructions and increases code size, but avoids the performance
impact of a hard barrier.
Without the -mtrack-speculation option, __builtin_speculation_safe_value
expands to a
ISB
DSB SY
sequence after a conditional branch, before the
speculation safe variable is used. With -mtrack-speculation a
CSEL tracker, tracker, XZR, [cond];
AND safeval,tracker;
CSDB
sequence is added instead, clearing the vulnerable variable by
AND'ing it with the tracker register, which is zero during speculative
execution. [cond] are the status flags which will only be true during
speculative execution. For more information on
__builtin_speculation_safe_value and the -mtrack-speculation option see
https://developer.arm.com/support/arm-security-updates/speculative-processor-vulnerability/compiler-support-for-mitigations
The -mtracking option was not added, as the performance impact of the
mitigation is low, and there is only one occurence.
Change-Id: Ic9e66d1f4a5155e42e3e4055594974c230bfba3c
Signed-off-by: Joel Hutton <Joel.Hutton@Arm.com>
Some platforms can only resume from system suspend from the boot
CPU, hence they should only enter that state from that same core.
The following commit presents an interface that allows the platform to
reject system suspend entry near its very last stage (last CPU).
When a platform enables its caches before it accesses the
psci_non_cpu_pd_nodes structure then explicit cache maintenance
is not needed.
Signed-off-by: Andrew F. Davis <afd@ti.com>
The MMU is not disabled in this path, update the comment to
reflect this. Also clarify that both paths call prepare_cpu_pwr_dwn(),
but the second path does stack cache maintenance.
Signed-off-by: Andrew F. Davis <afd@ti.com>
Mark the initialization functions in BL31, such as context management,
EHF, RAS and PSCI as __init so that they can be reclaimed by the
platform when no longer needed
Change-Id: I7446aeee3dde8950b0f410cb766b7a2312c20130
Signed-off-by: Daniel Boulby <daniel.boulby@arm.com>
During cold boot, the initial translation tables are created with data
caches disabled, so all modifications go to memory directly. After the
MMU is enabled and data cache is enabled, any modification to the tables
goes to data cache, and eventually may get flushed to memory.
If CPU0 modifies the tables while CPU1 is off, CPU0 will have the
modified tables in its data cache. When CPU1 is powered on, the MMU is
enabled, then it enables coherency, and then it enables the data cache.
Until this is done, CPU1 isn't in coherency, and the translation tables
it sees can be outdated if CPU0 still has some modified entries in its
data cache.
This can be a problem in some cases. For example, the warm boot code
uses only the tables mapped during cold boot, which don't normally
change. However, if they are modified (and a RO page is made RW, or a XN
page is made executable) the CPU will see the old attributes and crash
when it tries to access it.
This doesn't happen in systems with HW_ASSISTED_COHERENCY or
WARMBOOT_ENABLE_DCACHE_EARLY. In these systems, the data cache is
enabled at the same time as the MMU. As soon as this happens, the CPU is
in coherency.
There was an attempt of a fix in psci_helpers.S, but it didn't solve the
problem. That code has been deleted. The code was introduced in commit
<264410306381> ("Invalidate TLB entries during warm boot").
Now, during a map or unmap operation, the memory associated to each
modified table is flushed. Traversing a table will also flush it's
memory, as there is no way to tell in the current implementation if the
table that has been traversed has also been modified.
Change-Id: I4b520bca27502f1018878061bc5fb82af740bb92
Signed-off-by: Antonio Nino Diaz <antonio.ninodiaz@arm.com>
If either USE_COHERENT_MEM or HW_ASSISTED_COHERENCY being true
should cause us to not enter the ifdef block, then the logic
is not correct here. Posibly bad use of De Morgan's law?
Fix this.
Signed-off-by: Andrew F. Davis <afd@ti.com>
Use a _ prefix for Macro arguments to prevent that argument from
hiding variables of the same name in the outer scope
Rule 5.3: An identifier declared in an inner scope shall not
hide an identifier declared in an outer scope
Fixed For:
make LOG_LEVEL=50 PLAT=fvp
Change-Id: I67b6b05cbad4aeca65ce52981b4679b340604708
Signed-off-by: Daniel Boulby <daniel.boulby@arm.com>
Initializes each element of the last_cpu_in_non_cpu_pd array in PSCI
stat implementation to -1, the reset value. This satisfies MISRA rule
9.3.
Previously, only the first element of the array was initialized to -1.
Change-Id: I666c71e6c073710c67c6d24c07a219b1feb5b773
Signed-off-by: Jonathan Wright <jonathan.wright@arm.com>
When the source code says 'SMCC' it is talking about the SMC Calling
Convention. The correct acronym is SMCCC. This affects a few definitions
and file names.
Some files have been renamed (smcc.h, smcc_helpers.h and smcc_macros.S)
but the old files have been kept for compatibility, they include the
new ones with an ERROR_DEPRECATED guard.
Change-Id: I78f94052a502436fdd97ca32c0fe86bd58173f2f
Signed-off-by: Antonio Nino Diaz <antonio.ninodiaz@arm.com>
Rule 8.3: All declarations of an object or function shall
use the same names and type qualifiers.
Change-Id: Iff384187c74a598a4e73f350a1893b60e9d16cec
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
During the warm boot sequence:
1. The MMU is enabled with the data cache disabled. The MMU table walker
is set up to access the translation tables as in cacheable memory,
but its accesses are non-cacheable because SCTLR_EL3.C controls them
as well.
2. The interconnect is set up and the CPU enters coherency with the
rest of the system.
3. The data cache is enabled.
If the support for dynamic translation tables is enabled and another CPU
makes changes to a region, the changes may only be present in the data
cache, not in RAM. The CPU that is booting isn't in coherency with the
rest of the system, so the table walker of that CPU isn't either. This
means that it may read old entries from RAM and it may have invalid TLB
entries corresponding to the dynamic mappings.
This is not a problem for the boot code because the mapping is 1:1 and
the regions are static. However, the code that runs after the boot
sequence may need to access the dynamically mapped regions.
This patch invalidates all TLBs during warm boot when the dynamic
translation tables support is enabled to prevent this problem.
Change-Id: I80264802dc0aa1cb3edd77d0b66b91db6961af3d
Signed-off-by: Antonio Nino Diaz <antonio.ninodiaz@arm.com>
On some platforms it may be necessary to discover the SMCCC version
via a PSCI features call.
Change-Id: I95281ac2263ca9aefda1809eb03464fbdb8ac24d
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
The suspend hook is published at the start of a CPU powerdown
operation. The resume hook is published at the end of a CPU powerup
operation.
Change-Id: I50c05e2dde0d33834095ac41b4fcea4c161bb434
Signed-off-by: Dimitris Papastamos <dimitris.papastamos@arm.com>
There is an edge case where the cache maintaince done in
psci_do_cpu_off may not seen by some cores. This case is handled in
psci_cpu_on_start but it hasn't handled in psci_affinity_info.
Change-Id: I4d64f3d1ca9528e364aea8d04e2d254f201e1702
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
If ARMv7 based platform does not set ARM_CORTEX_Ax=yes, platform
shall define ARMV7_SUPPORTS_GENERIC_TIMER to enable generic timer
support.
Signed-off-by: Etienne Carriere <etienne.carriere@linaro.org>
This patch includes various fixes for PSCI STAT functionality
relating to timestamp collection:
1. The PSCI stat accounting for retention states for higher level
power domains were done outside the locks which could lead to
spurious values in some race conditions. This is moved inside
the locks. Also, the call to start the stat accounting was redundant
which is now removed.
2. The timestamp wrap-around case when calculating residency did
not cater for AArch32. This is now fixed.
3. In the warm boot path, `plat_psci_stat_accounting_stop()` was
getting invoked prior to population of target power states. This
is now corrected.
Change-Id: I851526455304fb74ff0a724f4d5318cd89e19589
Signed-off-by: Soby Mathew <soby.mathew@arm.com>
This allows other EL3 components to subscribe to CPU on events.
Update Firmware Design guide to list psci_cpu_on_finish as an available
event.
Change-Id: Ida774afe0f9cdce4021933fcc33a9527ba7aaae2
Signed-off-by: Jeenu Viswambharan <jeenu.viswambharan@arm.com>
With this patch the PSCI_FEATURES API correctly reports availability
of the PSCI_MEM_PROTECT_CHECK API - PSCI_MEM_CHK_RANGE_AARCH64 is
added to the PSCI capabilities mask, PSCI_CAP_64BIT_MASK
Change-Id: Ic90ee804deaadf0f948dc2d46ac5fe4121ef77ae
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
This patch implements PSCI_SYSTEM_RESET2 API as defined in PSCI
v1.1 specification. The specification allows architectural and
vendor-specific resets via this API. In the current specification,
there is only one architectural reset, the warm reset. This reset is
intended to provide a fast reboot path that guarantees not to reset
system main memory.
Change-Id: I057bb81a60cd0fe56465dbb5791d8e1cca025bd3
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
This patch adds the generic code that links the psci smc handler
with the platform function that implements the mem_protect and
mem_check_range functionalities. These functions are optional
APIs added in PSCI v1.1 (ARM DEN022D).
Change-Id: I3bac1307a5ce2c7a196ace76db8317e8d8c8bb3f
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
psci_set_power_off_state only initializes a local variable, so there
isn't any reason why it should be done while the lock is held.
Change-Id: I1c62f4cd5d860d102532e5a5350152180d41d127
Signed-off-by: Roberto Vargas <roberto.vargas@arm.com>
Previously commit 555ebb34db8f3424c1b394df2f10ecf9c1f70901 attmpted to fix this
GCC issue:
services/std_svc/psci/psci_common.c: In function 'psci_do_state_coordination':
services/std_svc/psci/psci_common.c:220:27: error: array subscript is above
array bounds [-Werror=array-bounds]
psci_req_local_pwr_states[pwrlvl - 1][cpu_idx] = req_pwr_state;
This fix doesn't work as asserts aren't built in non-debug build flows.
Let's use GCCs #pragma option (documented here:
https://gcc.gnu.org/onlinedocs/gcc/Diagnostic-Pragmas.html) to avoid
this false positive instead.
Signed-off-by: Alistair Francis <alistair.francis@xilinx.com>
This patch adds an early suspend handler, that executes with
SMP and data cache enabled. This handler allows platforms to
perform any early actions during the CPU suspend entry sequence.
This handler is optional and platforms can choose to implement it
depending on their needs. The `pwr_domain_suspend` handler still
exists and platforms can keep on using it without any side effects.
Signed-off-by: Varun Wadekar <vwadekar@nvidia.com>
Madhukar Pappireddy
Soby Mathew
Alexei Fedorov
Deepika Bhavnani
Julius Werner
Andrew F. Davis
Antonio Nino Diaz
Joel Hutton
ldts
Daniel Boulby
Roberto Vargas
Dimitris Papastamos
Jonathan Wright
davidcunado-arm
Etienne Carriere
Jeenu Viswambharan
Alistair Francis
Varun Wadekar